Chronic obstructive pulmonary disease (COPD) will soon become the third leading cause of death in the world. COPD is caused by multiple factors including environmental exposures, infections, inflammation, and genetic predisposition. Several lines of investigation have indicated that, in advanced disease, increased bacterial colonization leads to airway inflammation and accelerated airway obstruction. However, in early COPD, culture-dependent techniques commonly failed to isolate bacteria. Thus, the role of microbes in COPD has been difficult to study. We hypothesize that microbes are part of the pathogenesis of early COPD, contributing to inflammatory processes. This proposal will use culture independent techniques to evaluate the contribution of the microbiome to airway inflammation in early COPD. It will also investigate the microbial genomic potential (metagenome) and metabolic profile (metabolome) to better understand functional aspects of the lung microbiome. Aim 1 will assess lung microbiome and inflammation in smokers with and without COPD. We will perform research bronchoscopy in 48 subjects with mild-moderate COPD and 48 matched smoker controls to measure lung inflammation with multiplexed cytokine assays of bronchoalveolar lavage and lower airway bacteria by 16S rRNA gene sequencing. Aim 2 will investigate active bacterial metabolic pathways in the lower airways using metagenomics and metabolomics by using the samples obtained in AIM 1. We will test whether differences in genomic potential of the lower airway microbiome accounts for variation in the lung metabolome. Presence of correlation between operons encoding synthetic pathways and the metabolic products of those pathways would support that the lung microbiome is metabolically active, thus composed by live microorganisms. Aim 3 will assess relationships among different inter-omic datasets (microbiome, metagenome, metabolome and host immune responses) in COPD and controls with the goal of integrating information from vastly different platforms to achieve a multidimensional model of disease. Under the mentorship of Drs. Blaser, Weiden and with the collaboration of Dr. Virgin, Fiehn, and Clemente this research plan will serve as the foundation of a training plan that uses state-of-the-art molecular and bio-informatic methods to improve our understanding of complex bacterial community networks and their association with host immune response in COPD.